Baldor VFD Bus Bar Fault

[jscaufi]

I have an industrial customer that suffered damage to a 200 hp Baldor 18H Vector Drive. A worker disconnected power to the system, and then about a minute later the plant tripped off. Our substation recloser control saw an OCR (breaker) alarm for min trip level of the OCR, but it didn't trip. This occurred twice, and interviews with the workers revealed the plant tripped twice. Other than that, it was a very quiet day for the utility, with only 2 entries in the log for the entire state we serve (trouble calls in other words).

When power was reset to the plant (I believe a ground fault relay tripped the main breaker), and the tech reapplied power to the drive that had no load on it, just idling, it arced inside, then the upstream breaker smoked. They are unsure if these occurred simultaneously, or one smoked before the other.

Opening the case of the drive revealed severe arcing and pitting of the L1 bus bar at a corner of the metal chassis were it comes within 1/2" of the metal bus bar. The metal bus bar suffered the most arcing to the point it is pitted about 1/8" deep. There are also other signs of arcing throughout the chassis and components. The upstream breaker was a refurbished breaker. It has smoke marks coming out the side seams.

The drive guys says this has never happened with any of their thousands of drives of this model. I compared with an identical drive adjacent to this one and the clearance is a little wider at the point the bad one arced. However, air insulation flashover distances at that voltage are about 0.07 inches. Should be plenty of clearance. Area was a little dusty but not significant.

They brought us in to monitor for a while, which I have commenced. Nothing severe in first 24 hrs. Nothing appeared to get across those contact points, no dead insects, animals, or foreign material on bottom of the case.

I don't see how utility power could do this type of specific damage so limited to one customer, and limited to one breaker and VFD. Previous monitoring for other problems revealed nothing unusual. There are input and output line reactors on each of the two 200 hp drives. Similar problems have intermittently affected these two drives over the past 3 years but nothing this severe. Always seems to happen on weekends, and the problems alternate between drives. A few days prior, we had severe windstorms but it was 11 mph the morning this happened.

Any ideas or suggestions?

thanks

jack

 

[sreid]

I'm thinking a metalic arc. A small piece of wire or metal chip falls across the bus and the inverter cabinet (ground). The metal vaporizes and starts a metalic arc which has the tempeature and heat of arc welding. No trace of the metal chip is likely to be found. This would definately cause pitting and cause a ground fault.

 

[mc5w]

The drive's time came as mentioned above. I have had an instance where the neutral point of a wye motor was ground faulted due to a conduit box connector contacting a ring terminal in the motor junction box but that should trip the drive, not burn it up.

 

[jscaufi]

Thanks sreid and mc5w. We thought about that foreign object getting across that point and vaporizing, but it is 1/4 the distance from the top of the cabinet, the cabinet was sealed (screwed) shut, and the only thing above that point was a couple of coils or small transformers with coils wrapped in paper insulation.

I have also learned a few more details.

Worker pressed disable button that stops machine where the motor is located that is powered by the drive. Worker then opened disconnect between output line reactor of and machine where motor is located. This disconnect is close to machine and motor. 1-2 mins after he went to work on the machine, power to plant went out. About 5 mn after plant tripped off, maintenance worker arrive at main panel where he saw main breaker tripped. Room was filled with smoke. They shut off every breaker fed off main breaker panel, then re-energized main switch, and turned on all breakers one at a time. When they tried to reset the breaker that fed the drive and machine he was working on, the breaker flashed and the main switch tripped a 2nd time.

They went through the procedure again, but did not try that breaker that flashed previously. A plant engineer arrived later to look at the drive and found the arcing.

The drive tripping (not the arcing) has affected two identical drives and seems to be only on weekends when a particular shift comes on to turn on the machine that had been idle. The trips manifest by the breaker tripping. No one has ever figured out why this happens, but now it has caused damage.

 

[weh3]

Is the utility switching any equipment nearby, especially capacitor banks? Switching transient voltages will evade o/c relaying and your protective device coordination.

William

 

[jscaufi]

No utility banks that affect this location. However I have never heard of a capacitor bank energizing transient being able to start an arc from the limits they usually hit. A typical capacitor bank energizing transient could be as low as negligible to upwards of 1.9 per unit. 1.9 per unit is the largest I have ever captured on our monitoring equipment, and that was at a location with very light load, very near the high voltage substation where the large capacitor bank was located. Even if we hit 2 per unit, and that made the line-ground RMS voltage hit 2*277 volts or 783 volts peak, that is still within the 750 V to 2kV volt flashover range, which has a corresponding flashover distance of 0.19 inches. This bus bar was about 1/2 to 5/8th inch from the arc point of the chassis. The voltage transient would have to be around 2 kV to cause flashover at about 1.5 inches.

I realize a cap bank transient can cause the drives to trip, and have investigated many of these cases, but that is not the problem here.

They do have an auto cap bank for PF correction on the main panel however.

j

 

[castera]

When the original fault occured,how was power removed from the drive. Was the drive stopped in the correct manner or was the input power disconnected from the drive ?.
It sounds to me like some kind of bus over voltage may have occured.
Did they kill the drive before the load had come to rest?.
Does the drive Regen back into the supply or via a resistor ?.
Is the regen working correctly, is the regen resitor in good order.